These studies examine how hydrophobic flanking sequences shape the process-ing of peptide antigens by antigen presenting cells (APC). In the past year, we focused on antigen processing and presentation of lemA, a bacterial polypeptide produced by Listeria monocytogenes (LM). These studies have now been extended using recombinant antigen products to address the impact of these flanking structures on the processing of other peptide antigens presented by conventional class Ia MHC products. LemA is a hydrophobic 185 amino-acid polypeptide of unknown function, and its 6 N-terminal amino acids bind to the nonclassical class I MHC product H2M3. The resulting complex elicits a brisk CD8 response in LM infected mice in vivo. We initially became interested in lemA because it has several unusual properties. The immunogenic core of the molecule is profoundly resistant to degradation by a variety of proteolytic enzymes including proteinase K. Yet, unlike most class I-dependent antigens which can not be presented to CD8 as an exogenous antigen, lemA is processed by a variety of APC via a novel brefeldin- and pepstatin-sensitive, TAP-independent pathway. Using molecular techniques, we have constructed and expressed a truncated lemA product, containing only the immunogenic amino-terminus and adjacent transmembrane portion of lemA with a carboxyterminal histidine tag to facilitate purification. The resulting small (46 amino-acid) construct retains all the distinctive antigenic properties of native lemA described above in the absence of other bacterial proteins or lipids. We have subsequently placed other immunogenic amino acid sequences adjacent to the hydrophobic structural portion of this construct, and demonstrated that the resulting chimeric molecules can be processed by APC as well. Processing is mediated in part by the brefeldin-sensitive pathway described above, however we note promiscuous processing via alternative cytoplasmic and post-Golgi pathways as well. Using T cell hyridomas and antibodies specific for the complex formed by the binding of one of these peptides (SIINFEKL) to its presenting MHC product (H2Kb) we are in the process of analyzing where and how antigen presentation occurs in various types of APC. During the coming year, we hope: 1) to assess the efficacy of chimeric constructs (with and without adjuvants) in inducing a CD8 response directed against the model peptide SIINFEKL bound to H2Kb. These studies will be performed initially using freshly obtained lymphoid tissues ex vivo, but ultimately we will use immunofluorescent tetramers and elispot techniques to facilitate quanti-tative analysis in vivo; 2) We will also use fluorescently labeled CD8 cells from SIINFEKL-H2Kb transgenic mice to monitor the distribution and time course of immunogenic complexes in vivo in chimer-treated animals. Using these approaches, we hope to develop a rational method for assessing the potency and durability of antigen constructs in vivo, and to apply our findings in screening antigen constructs suitable for use in a vaccine setting.